Training & fine-tuning
Megatron-LM / Megatron Core
Separates scalable reference training scripts from reusable transformer and parallelism components.
Overview
Research summary
The Megatron-LM repository contains reference training scripts and Megatron Core, a library of GPU-oriented transformer building blocks. It supports combinations of tensor, pipeline, data, expert and context parallelism, alongside mixed-precision training. Research teams can use the provided recipes, while framework developers can integrate the lower-level components into custom training systems. The repository’s main license is BSD-3-Clause with separately identified Apache-2.0 and MIT portions; the README license badge alone does not describe that full scope.
Repository summary
- Stars
- 18,051
- Open issues
- Unavailable
- Last push
- 2026-10-02
- Commits, 90 days
- Unavailable
- Repository activity
- Not scored
- Version
- Unavailable
Recorded catalogue figures. View repository data and provenance →
Classification
Pricing & services
Paid services unknown
Whether the provider offers paid products or services has not been established.
Licence scope
The top-level LICENSE defines the default and separately identifies differently licensed portions; these are component scopes, not alternatives for every file.
Implementation
Recorded implementation details and interfaces for Megatron-LM / Megatron Core.
Implementation details
Python library and supporting tools
- Languages
- Python
- Repository type
- source
Recorded interfaces and capabilities
Licence scope
The top-level LICENSE defines the default and separately identifies differently licensed portions; these are component scopes, not alternatives for every file.
Repository
Repository snapshots, release information and recorded maintenance signals.
Repository snapshot
- Stars
- 18,051
- Open issues
- Unavailable
- Last push
- 2026-10-02
- Commits, 90 days
- Unavailable
- Repository activity
- Not scored
- Archived
- Not recorded
Repository activity is a snapshot, not a quality or popularity ranking. It combines recent-push freshness (50%), 90-day commits (30%) and issue pressure (20%).
Maintenance and provenance
- Catalogue snapshot
- 2026-10-06
- Stars source
- Recorded fallback
- Last push source
- Recorded fallback
Documentation
Recorded references and research provenance for this entry.
Recorded sources 5
- https://docs.nvidia.com/megatron-core/developer-guide/latest/get-started/quickstart.html Project page · Documentation
- https://github.com/NVIDIA/Megatron-LM Linked repository · Research reference
- https://api.github.com/repos/NVIDIA/Megatron-LM Research reference
- https://github.com/NVIDIA/Megatron-LM/blob/main/README.md Research reference
- https://github.com/NVIDIA/Megatron-LM/blob/main/LICENSE Research reference
Research metadata
- Research date
- 2026-10-02
- Catalogue snapshot
- 2026-10-06
Community & social 1 channels
Communities
- GitHub Discussionsgithub.com/NVIDIA/Megatron-LM/discussionsOfficial
Link details for GitHub Discussions
- Last checked